Kong times out by sending requests to stale upstream ip's

Hello All,

We are seeing timeouts, as kong-ingress-controller is failing to make put requests to update the upstreams and kong trying to send requests to pod ip’s that doesn’t exist anymore. This issue started happening after updating kong-ingress-controller (2.0.4), any inputs would be helpful

time="2023-01-12T00:28:44Z" level=error msg="could not update kong admin" error="4 errors occurred:\n\twhile processing event: {Create} upstream example.sample.8011.svc failed: HTTP status 409 (message: "UNIQUE violation detected on '{name=\"example.sample.8011.svc\"}'")